The field of human genetics has been transformed by genome-wide association studies (GWAS), which have identified thousands of genomic loci that influence hundreds of diseases and traits. In order to interpret the mechanisms by which these loci act, consortia and individual investigators have generated hundreds of functional genomic data sets with the goal of identifying important gene regulatory elements. This proposal describes a set of novel methods for combining hundreds of functional genomics datasets with a large GWAS of schizophrenia to identify the precise cell types, pathways, and gene regulatory elements most relevant to disease. In Aim 1, we describe a statistical method for fine-mapping risk loci using functional genomic information. We propose to systematically scan through thousands of genomic annotations to identify those that are most important for understanding schizophrenia risk. In Aim 2, we propose to combine GWAS data with gene expression data by jointly analyzing association studies of traits with association studies of gene expression (expression quantitative trait loci). Finally, in Aim 3, we propose to extend these methods to analyze multiple GWAS at once, with a specific emphasis on inferring the causal relationships between traits. The methods developed in this proposal will be widely application in studies of disease, and the analyses performed will generate direct predictions about the molecular consequences of alleles that influence schizophrenia risk.